From e8b26b2135dedc0284490bfeac06dfc4418d0105 Mon Sep 17 00:00:00 2001 From: Artemy Kovalyov Date: Tue, 19 Mar 2019 11:24:38 +0200 Subject: [PATCH 01/15] net/mlx5: Decrease default mr cache size Delete initialization of high order entries in mr cache to decrease initial memory footprint. When required, the administrator can populate the entries with memory keys via the /sys interface. This approach is very helpful to significantly reduce the per HW function memory footprint in virtualization environments such as SRIOV. Fixes: 9603b61de1ee ("mlx5: Move pci device handling from mlx5_ib to mlx5_core") Signed-off-by: Artemy Kovalyov Signed-off-by: Moni Shoua Signed-off-by: Leon Romanovsky Reported-by: Shalom Toledo Acked-by: Or Gerlitz Signed-off-by: Saeed Mahameed --- .../net/ethernet/mellanox/mlx5/core/main.c | 20 ------------------- 1 file changed, 20 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c index 70cc906a102b2..76716419370df 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/main.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c @@ -164,26 +164,6 @@ static struct mlx5_profile profile[] = { .size = 8, .limit = 4 }, - .mr_cache[16] = { - .size = 8, - .limit = 4 - }, - .mr_cache[17] = { - .size = 8, - .limit = 4 - }, - .mr_cache[18] = { - .size = 8, - .limit = 4 - }, - .mr_cache[19] = { - .size = 4, - .limit = 2 - }, - .mr_cache[20] = { - .size = 4, - .limit = 2 - }, }, }; From bc87a0036826a37b43489b029af8143bd07c6cca Mon Sep 17 00:00:00 2001 From: Gavi Teitz Date: Mon, 11 Mar 2019 11:56:34 +0200 Subject: [PATCH 02/15] net/mlx5e: Fix error handling when refreshing TIRs Previously, a false positive would be caught if the TIRs list is empty, since the err value was initialized to -ENOMEM, and was only updated if a TIR is refreshed. This is resolved by initializing the err value to zero. Fixes: b676f653896a ("net/mlx5e: Refactor refresh TIRs") Signed-off-by: Gavi Teitz Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_common.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c index 3078491cc0d06..8100786f6fb5c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c @@ -141,15 +141,17 @@ int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb) { struct mlx5_core_dev *mdev = priv->mdev; struct mlx5e_tir *tir; - int err = -ENOMEM; + int err = 0; u32 tirn = 0; int inlen; void *in; inlen = MLX5_ST_SZ_BYTES(modify_tir_in); in = kvzalloc(inlen, GFP_KERNEL); - if (!in) + if (!in) { + err = -ENOMEM; goto out; + } if (enable_uc_lb) MLX5_SET(modify_tir_in, in, ctx.self_lb_block, From 8998576bd9c695ef1297540a50d7b3abbc69286b Mon Sep 17 00:00:00 2001 From: Dmytro Linkin Date: Mon, 4 Feb 2019 09:45:47 +0000 Subject: [PATCH 03/15] net/mlx5e: Allow IPv4 ttl & IPv6 hop_limit rewrite for all L4 protocols For some protocols we are not allowing IP header rewrite offload, since the HW is not capable to properly adjust the l4 checksum. However, TTL & HOPLIMIT modification can be done for all IP protocols, because they are not part of the pseudo header taken into account for checksum. Fixes: 738678817573 ("drivers: net: use flow action infrastructure") Signed-off-by: Dmytro Linkin Signed-off-by: Saeed Mahameed --- .../net/ethernet/mellanox/mlx5/core/en_tc.c | 52 +++++++++++++++++-- 1 file changed, 48 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index b4967a0ff8c7b..2b85f93a1c52b 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -2158,6 +2158,52 @@ static bool csum_offload_supported(struct mlx5e_priv *priv, return true; } +struct ip_ttl_word { + __u8 ttl; + __u8 protocol; + __sum16 check; +}; + +struct ipv6_hoplimit_word { + __be16 payload_len; + __u8 nexthdr; + __u8 hop_limit; +}; + +static bool is_action_keys_supported(const struct flow_action_entry *act) +{ + u32 mask, offset; + u8 htype; + + htype = act->mangle.htype; + offset = act->mangle.offset; + mask = ~act->mangle.mask; + /* For IPv4 & IPv6 header check 4 byte word, + * to determine that modified fields + * are NOT ttl & hop_limit only. + */ + if (htype == FLOW_ACT_MANGLE_HDR_TYPE_IP4) { + struct ip_ttl_word *ttl_word = + (struct ip_ttl_word *)&mask; + + if (offset != offsetof(struct iphdr, ttl) || + ttl_word->protocol || + ttl_word->check) { + return true; + } + } else if (htype == FLOW_ACT_MANGLE_HDR_TYPE_IP6) { + struct ipv6_hoplimit_word *hoplimit_word = + (struct ipv6_hoplimit_word *)&mask; + + if (offset != offsetof(struct ipv6hdr, payload_len) || + hoplimit_word->payload_len || + hoplimit_word->nexthdr) { + return true; + } + } + return false; +} + static bool modify_header_match_supported(struct mlx5_flow_spec *spec, struct flow_action *flow_action, u32 actions, @@ -2165,9 +2211,9 @@ static bool modify_header_match_supported(struct mlx5_flow_spec *spec, { const struct flow_action_entry *act; bool modify_ip_header; - u8 htype, ip_proto; void *headers_v; u16 ethertype; + u8 ip_proto; int i; if (actions & MLX5_FLOW_CONTEXT_ACTION_DECAP) @@ -2187,9 +2233,7 @@ static bool modify_header_match_supported(struct mlx5_flow_spec *spec, act->id != FLOW_ACTION_ADD) continue; - htype = act->mangle.htype; - if (htype == FLOW_ACT_MANGLE_HDR_TYPE_IP4 || - htype == FLOW_ACT_MANGLE_HDR_TYPE_IP6) { + if (is_action_keys_supported(act)) { modify_ip_header = true; break; } From 8e949363f017e2011464812a714fb29710fb95b4 Mon Sep 17 00:00:00 2001 From: Aditya Pakki Date: Tue, 19 Mar 2019 16:42:40 -0500 Subject: [PATCH 04/15] net: mlx5: Add a missing check on idr_find, free buf idr_find() can return a NULL value to 'flow' which is used without a check. The patch adds a check to avoid potential NULL pointer dereference. In case of mlx5_fpga_sbu_conn_sendmsg() failure, free buf allocated using kzalloc. Fixes: ab412e1dd7db ("net/mlx5: Accel, add TLS rx offload routines") Signed-off-by: Aditya Pakki Reviewed-by: Yuval Shaia Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c index 5cf5f2a9d51fe..8de64e88c670c 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/tls.c @@ -217,15 +217,21 @@ int mlx5_fpga_tls_resync_rx(struct mlx5_core_dev *mdev, u32 handle, u32 seq, void *cmd; int ret; + rcu_read_lock(); + flow = idr_find(&mdev->fpga->tls->rx_idr, ntohl(handle)); + rcu_read_unlock(); + + if (!flow) { + WARN_ONCE(1, "Received NULL pointer for handle\n"); + return -EINVAL; + } + buf = kzalloc(size, GFP_ATOMIC); if (!buf) return -ENOMEM; cmd = (buf + 1); - rcu_read_lock(); - flow = idr_find(&mdev->fpga->tls->rx_idr, ntohl(handle)); - rcu_read_unlock(); mlx5_fpga_tls_flow_to_cmd(flow, cmd); MLX5_SET(tls_cmd, cmd, swid, ntohl(handle)); @@ -238,6 +244,8 @@ int mlx5_fpga_tls_resync_rx(struct mlx5_core_dev *mdev, u32 handle, u32 seq, buf->complete = mlx_tls_kfree_complete; ret = mlx5_fpga_sbu_conn_sendmsg(mdev->fpga->tls->conn, buf); + if (ret < 0) + kfree(buf); return ret; } From 80a2a9026b24c6bd34b8d58256973e22270bedec Mon Sep 17 00:00:00 2001 From: Yuval Avnery Date: Mon, 11 Mar 2019 06:18:24 +0200 Subject: [PATCH 05/15] net/mlx5e: Add a lock on tir list Refresh tirs is looping over a global list of tirs while netdevs are adding and removing tirs from that list. That is why a lock is required. Fixes: 724b2aa15126 ("net/mlx5e: TIRs management refactoring") Signed-off-by: Yuval Avnery Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_common.c | 7 +++++++ include/linux/mlx5/driver.h | 2 ++ 2 files changed, 9 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c index 8100786f6fb5c..1539cf3de5dc9 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_common.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_common.c @@ -45,7 +45,9 @@ int mlx5e_create_tir(struct mlx5_core_dev *mdev, if (err) return err; + mutex_lock(&mdev->mlx5e_res.td.list_lock); list_add(&tir->list, &mdev->mlx5e_res.td.tirs_list); + mutex_unlock(&mdev->mlx5e_res.td.list_lock); return 0; } @@ -53,8 +55,10 @@ int mlx5e_create_tir(struct mlx5_core_dev *mdev, void mlx5e_destroy_tir(struct mlx5_core_dev *mdev, struct mlx5e_tir *tir) { + mutex_lock(&mdev->mlx5e_res.td.list_lock); mlx5_core_destroy_tir(mdev, tir->tirn); list_del(&tir->list); + mutex_unlock(&mdev->mlx5e_res.td.list_lock); } static int mlx5e_create_mkey(struct mlx5_core_dev *mdev, u32 pdn, @@ -114,6 +118,7 @@ int mlx5e_create_mdev_resources(struct mlx5_core_dev *mdev) } INIT_LIST_HEAD(&mdev->mlx5e_res.td.tirs_list); + mutex_init(&mdev->mlx5e_res.td.list_lock); return 0; @@ -159,6 +164,7 @@ int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb) MLX5_SET(modify_tir_in, in, bitmask.self_lb_en, 1); + mutex_lock(&mdev->mlx5e_res.td.list_lock); list_for_each_entry(tir, &mdev->mlx5e_res.td.tirs_list, list) { tirn = tir->tirn; err = mlx5_core_modify_tir(mdev, tirn, in, inlen); @@ -170,6 +176,7 @@ int mlx5e_refresh_tirs(struct mlx5e_priv *priv, bool enable_uc_lb) kvfree(in); if (err) netdev_err(priv->netdev, "refresh tir(0x%x) failed, %d\n", tirn, err); + mutex_unlock(&mdev->mlx5e_res.td.list_lock); return err; } diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h index 022541dc5dbfd..0d07296488448 100644 --- a/include/linux/mlx5/driver.h +++ b/include/linux/mlx5/driver.h @@ -594,6 +594,8 @@ enum mlx5_pagefault_type_flags { }; struct mlx5_td { + /* protects tirs list changes while tirs refresh */ + struct mutex list_lock; struct list_head tirs_list; u32 tdn; }; From 8d047bf56a2cc13d90e6a5074015d65045fd43e7 Mon Sep 17 00:00:00 2001 From: Aya Levin Date: Thu, 28 Feb 2019 09:27:33 +0200 Subject: [PATCH 06/15] net/mlx5: ethtool, Fix type analysis of advertised link-mode Ethtool option set_link_ksettings allows setting of legacy link-modes or extended link-modes. Refine the decision of which type of link-modes is set. Fixes: 6a897372417e ("net/mlx5: ethtool, Add ethtool support for 50Gbps per lane link modes") Signed-off-by: Aya Levin Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index a0987cc5fe4a1..561e36af8e77d 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -997,8 +997,9 @@ int mlx5e_ethtool_set_link_ksettings(struct mlx5e_priv *priv, #define MLX5E_PTYS_EXT ((1ULL << ETHTOOL_LINK_MODE_50000baseKR_Full_BIT) - 1) - ext_requested = (link_ksettings->link_modes.advertising[0] > - MLX5E_PTYS_EXT); + ext_requested = !!(link_ksettings->link_modes.advertising[0] > + MLX5E_PTYS_EXT || + link_ksettings->link_modes.advertising[1]); ext_supported = MLX5_CAP_PCAM_FEATURE(mdev, ptys_extended_ethernet); /*when ptys_extended_ethernet is set legacy link modes are deprecated */ From dd1b9e09c12b4231148f446c2eefd886ef6e3ddd Mon Sep 17 00:00:00 2001 From: Aya Levin Date: Thu, 28 Feb 2019 09:39:02 +0200 Subject: [PATCH 07/15] net/mlx5: ethtool, Allow legacy link-modes configuration via non-extended ptys Allow configuration of legacy link-modes even when extended link-modes are supported. This requires reading of legacy advertisement even when extended link-modes are supported. Since legacy and extended advertisement are mutually excluded, wait for empty reply from extended advertisement before reading legacy advertisement. Fixes: 6a897372417e ("net/mlx5: ethtool, Add ethtool support for 50Gbps per lane link modes") Signed-off-by: Aya Levin Signed-off-by: Saeed Mahameed --- .../net/ethernet/mellanox/mlx5/core/en/port.c | 3 -- .../ethernet/mellanox/mlx5/core/en_ethtool.c | 47 ++++++++++++------- 2 files changed, 31 insertions(+), 19 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c index 122927f3a6005..d5e5afbdca6dc 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port.c @@ -96,9 +96,6 @@ int mlx5_port_query_eth_proto(struct mlx5_core_dev *dev, u8 port, bool ext, if (!eproto) return -EINVAL; - if (ext != MLX5_CAP_PCAM_FEATURE(dev, ptys_extended_ethernet)) - return -EOPNOTSUPP; - err = mlx5_query_port_ptys(dev, out, sizeof(out), MLX5_PTYS_EN, port); if (err) return err; diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c index 561e36af8e77d..5efce4a3ff796 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c @@ -603,16 +603,18 @@ static void ptys2ethtool_supported_link(struct mlx5_core_dev *mdev, __ETHTOOL_LINK_MODE_MASK_NBITS); } -static void ptys2ethtool_adver_link(struct mlx5_core_dev *mdev, - unsigned long *advertising_modes, - u32 eth_proto_cap) +static void ptys2ethtool_adver_link(unsigned long *advertising_modes, + u32 eth_proto_cap, bool ext) { unsigned long proto_cap = eth_proto_cap; struct ptys2ethtool_config *table; u32 max_size; int proto; - mlx5e_ethtool_get_speed_arr(mdev, &table, &max_size); + table = ext ? ptys2ext_ethtool_table : ptys2legacy_ethtool_table; + max_size = ext ? ARRAY_SIZE(ptys2ext_ethtool_table) : + ARRAY_SIZE(ptys2legacy_ethtool_table); + for_each_set_bit(proto, &proto_cap, max_size) bitmap_or(advertising_modes, advertising_modes, table[proto].advertised, @@ -794,12 +796,12 @@ static void get_supported(struct mlx5_core_dev *mdev, u32 eth_proto_cap, ethtool_link_ksettings_add_link_mode(link_ksettings, supported, Pause); } -static void get_advertising(struct mlx5_core_dev *mdev, u32 eth_proto_cap, - u8 tx_pause, u8 rx_pause, - struct ethtool_link_ksettings *link_ksettings) +static void get_advertising(u32 eth_proto_cap, u8 tx_pause, u8 rx_pause, + struct ethtool_link_ksettings *link_ksettings, + bool ext) { unsigned long *advertising = link_ksettings->link_modes.advertising; - ptys2ethtool_adver_link(mdev, advertising, eth_proto_cap); + ptys2ethtool_adver_link(advertising, eth_proto_cap, ext); if (rx_pause) ethtool_link_ksettings_add_link_mode(link_ksettings, advertising, Pause); @@ -854,8 +856,9 @@ static void get_lp_advertising(struct mlx5_core_dev *mdev, u32 eth_proto_lp, struct ethtool_link_ksettings *link_ksettings) { unsigned long *lp_advertising = link_ksettings->link_modes.lp_advertising; + bool ext = MLX5_CAP_PCAM_FEATURE(mdev, ptys_extended_ethernet); - ptys2ethtool_adver_link(mdev, lp_advertising, eth_proto_lp); + ptys2ethtool_adver_link(lp_advertising, eth_proto_lp, ext); } int mlx5e_ethtool_get_link_ksettings(struct mlx5e_priv *priv, @@ -872,6 +875,7 @@ int mlx5e_ethtool_get_link_ksettings(struct mlx5e_priv *priv, u8 an_disable_admin; u8 an_status; u8 connector_type; + bool admin_ext; bool ext; int err; @@ -886,6 +890,19 @@ int mlx5e_ethtool_get_link_ksettings(struct mlx5e_priv *priv, eth_proto_capability); eth_proto_admin = MLX5_GET_ETH_PROTO(ptys_reg, out, ext, eth_proto_admin); + /* Fields: eth_proto_admin and ext_eth_proto_admin are + * mutually exclusive. Hence try reading legacy advertising + * when extended advertising is zero. + * admin_ext indicates how eth_proto_admin should be + * interpreted + */ + admin_ext = ext; + if (ext && !eth_proto_admin) { + eth_proto_admin = MLX5_GET_ETH_PROTO(ptys_reg, out, false, + eth_proto_admin); + admin_ext = false; + } + eth_proto_oper = MLX5_GET_ETH_PROTO(ptys_reg, out, ext, eth_proto_oper); eth_proto_lp = MLX5_GET(ptys_reg, out, eth_proto_lp_advertise); @@ -899,7 +916,8 @@ int mlx5e_ethtool_get_link_ksettings(struct mlx5e_priv *priv, ethtool_link_ksettings_zero_link_mode(link_ksettings, advertising); get_supported(mdev, eth_proto_cap, link_ksettings); - get_advertising(mdev, eth_proto_admin, tx_pause, rx_pause, link_ksettings); + get_advertising(eth_proto_admin, tx_pause, rx_pause, link_ksettings, + admin_ext); get_speed_duplex(priv->netdev, eth_proto_oper, link_ksettings); eth_proto_oper = eth_proto_oper ? eth_proto_oper : eth_proto_cap; @@ -1001,16 +1019,13 @@ int mlx5e_ethtool_set_link_ksettings(struct mlx5e_priv *priv, MLX5E_PTYS_EXT || link_ksettings->link_modes.advertising[1]); ext_supported = MLX5_CAP_PCAM_FEATURE(mdev, ptys_extended_ethernet); - - /*when ptys_extended_ethernet is set legacy link modes are deprecated */ - if (ext_requested != ext_supported) - return -EPROTONOSUPPORT; + ext_requested &= ext_supported; speed = link_ksettings->base.speed; ethtool2ptys_adver_func = ext_requested ? mlx5e_ethtool2ptys_ext_adver_link : mlx5e_ethtool2ptys_adver_link; - err = mlx5_port_query_eth_proto(mdev, 1, ext_supported, &eproto); + err = mlx5_port_query_eth_proto(mdev, 1, ext_requested, &eproto); if (err) { netdev_err(priv->netdev, "%s: query port eth proto failed: %d\n", __func__, err); @@ -1038,7 +1053,7 @@ int mlx5e_ethtool_set_link_ksettings(struct mlx5e_priv *priv, if (!an_changes && link_modes == eproto.admin) goto out; - mlx5_port_set_eth_ptys(mdev, an_disable, link_modes, ext_supported); + mlx5_port_set_eth_ptys(mdev, an_disable, link_modes, ext_requested); mlx5_toggle_port_link(mdev); out: From 8a91ad9355c66c5026d3d911b434a25408ab876c Mon Sep 17 00:00:00 2001 From: Roi Dayan Date: Thu, 7 Mar 2019 09:27:18 +0200 Subject: [PATCH 08/15] net/mlx5: E-Switch, Fix access to invalid memory when toggling esw modes The esw fdb table has a union of legacy and offloads members. So if we were in a certain esw mode we could set some memebers and not set null which is fine as on destroy path and don't care. But then moving from legacy to switchdev a second time, the cleanup flow of legacy mode checks if a struct member was in use if it's not null so we need to make sure to reset the code to null when we init legacy mode. Fixes: 8da202b24913 ("net/mlx5: E-Switch, Add support for VEPA in legacy mode.") Signed-off-by: Roi Dayan Reviewed-by: Huy Nguyen Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index ecd2c747f7260..1a3b4d5d77fb4 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -431,6 +431,8 @@ static int esw_create_legacy_table(struct mlx5_eswitch *esw) { int err; + memset(&esw->fdb_table.legacy, 0, sizeof(struct legacy_fdb)); + err = esw_create_legacy_vepa_table(esw); if (err) return err; From 84be899f6fd233ff2aeaf14cc43e6457425122b2 Mon Sep 17 00:00:00 2001 From: Tonghao Zhang Date: Tue, 26 Feb 2019 04:28:32 -0800 Subject: [PATCH 09/15] net/mlx5e: Correctly use the namespace type when allocating pedit action The capacity of FDB offloading and NIC offloading table are different, and when allocating the pedit actions, we should use the correct namespace type. Fixes: c500c86b0c75d ("net/mlx5e: support for two independent packet edit actions") Cc: Pablo Neira Ayuso Signed-off-by: Tonghao Zhang Reviewed-by: Roi Dayan Acked-by: Pablo Neira Ayuso Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 2b85f93a1c52b..5fb5cab36bf65 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -2701,7 +2701,7 @@ static int parse_tc_fdb_actions(struct mlx5e_priv *priv, if (hdrs[TCA_PEDIT_KEY_EX_CMD_SET].pedits || hdrs[TCA_PEDIT_KEY_EX_CMD_ADD].pedits) { - err = alloc_tc_pedit_action(priv, MLX5_FLOW_NAMESPACE_KERNEL, + err = alloc_tc_pedit_action(priv, MLX5_FLOW_NAMESPACE_FDB, parse_attr, hdrs, extack); if (err) return err; From 5c1d260ed10cf08dd7a0299c103ad0a3f9a9f7a1 Mon Sep 17 00:00:00 2001 From: Roi Dayan Date: Thu, 21 Mar 2019 15:51:35 -0700 Subject: [PATCH 10/15] net/mlx5: E-Switch, Protect from invalid memory access in offload fdb table The esw offloads structures share a union with the legacy mode structs. Reset the offloads struct to zero in init to protect from null assumptions made by the legacy mode code. Signed-off-by: Roi Dayan Reviewed-by: Or Gerlitz Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c index f2260391be5b9..9b2d78ee22b88 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c @@ -1611,6 +1611,7 @@ static int esw_offloads_steering_init(struct mlx5_eswitch *esw, int nvports) { int err; + memset(&esw->fdb_table.offloads, 0, sizeof(struct offloads_fdb)); mutex_init(&esw->fdb_table.offloads.fdb_prio_lock); err = esw_create_offloads_fdb_tables(esw, nvports); From eca4a928585ac08147e5cc8e2111ecbc6279ee31 Mon Sep 17 00:00:00 2001 From: Omri Kahalon Date: Sun, 24 Feb 2019 16:31:08 +0200 Subject: [PATCH 11/15] net/mlx5: E-Switch, Fix esw manager vport indication for more vport commands Traditionally, the PF (Physical Function) which resides on vport 0 was the E-switch manager. Since the ECPF (Embedded CPU Physical Function), which resides on vport 0xfffe, was introduced as the E-Switch manager, the assumption that the E-switch manager is on vport 0 is incorrect. Since the eswitch code already uses the actual vport value, all we need is to always set other_vport=1. Signed-off-by: Omri Kahalon Reviewed-by: Max Gurtovoy Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index 1a3b4d5d77fb4..c938954599c90 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -105,8 +105,7 @@ static int arm_vport_context_events_cmd(struct mlx5_core_dev *dev, u16 vport, opcode, MLX5_CMD_OP_MODIFY_NIC_VPORT_CONTEXT); MLX5_SET(modify_nic_vport_context_in, in, field_select.change_event, 1); MLX5_SET(modify_nic_vport_context_in, in, vport_number, vport); - if (vport) - MLX5_SET(modify_nic_vport_context_in, in, other_vport, 1); + MLX5_SET(modify_nic_vport_context_in, in, other_vport, 1); nic_vport_ctx = MLX5_ADDR_OF(modify_nic_vport_context_in, in, nic_vport_context); @@ -134,8 +133,7 @@ static int modify_esw_vport_context_cmd(struct mlx5_core_dev *dev, u16 vport, MLX5_SET(modify_esw_vport_context_in, in, opcode, MLX5_CMD_OP_MODIFY_ESW_VPORT_CONTEXT); MLX5_SET(modify_esw_vport_context_in, in, vport_number, vport); - if (vport) - MLX5_SET(modify_esw_vport_context_in, in, other_vport, 1); + MLX5_SET(modify_esw_vport_context_in, in, other_vport, 1); return mlx5_cmd_exec(dev, in, inlen, out, sizeof(out)); } From 36acf63a066f7e095ef2322118c2742a177daa65 Mon Sep 17 00:00:00 2001 From: Huy Nguyen Date: Fri, 22 Mar 2019 09:42:08 -0500 Subject: [PATCH 12/15] net/mlx5: E-Switch, fix syndrome (0x678139) when turn on vepa Make sure the struct mlx5_flow_destination is zero before filling in the field. Fixes: 8da202b24913 ("net/mlx5: E-Switch, Add support for VEPA in legacy mode.") Signed-off-by: Huy Nguyen Reviewed-by: Daniel Jurgens Signed-off-by: Saeed Mahameed --- drivers/net/ethernet/mellanox/mlx5/core/eswitch.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c index c938954599c90..8a67fd197b792 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.c @@ -2157,6 +2157,7 @@ static int _mlx5_eswitch_set_vepa_locked(struct mlx5_eswitch *esw, /* Star rule to forward all traffic to uplink vport */ memset(spec, 0, sizeof(*spec)); + memset(&dest, 0, sizeof(dest)); dest.type = MLX5_FLOW_DESTINATION_TYPE_VPORT; dest.vport.num = MLX5_VPORT_UPLINK; flow_act.action = MLX5_FLOW_CONTEXT_ACTION_FWD_DEST; From 5ec983e924c7978aaec3cf8679ece9436508bb20 Mon Sep 17 00:00:00 2001 From: Huy Nguyen Date: Thu, 7 Mar 2019 14:49:50 -0600 Subject: [PATCH 13/15] net/mlx5e: Update xoff formula Set minimum speed in xoff threshold formula to 40Gbps Fixes: 0696d60853d5 ("net/mlx5e: Receive buffer configuration") Signed-off-by: Huy Nguyen Signed-off-by: Saeed Mahameed --- .../net/ethernet/mellanox/mlx5/core/en/port_buffer.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c index eac245a93f918..f00de0c987cd6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c @@ -122,7 +122,9 @@ static int port_set_buffer(struct mlx5e_priv *priv, return err; } -/* xoff = ((301+2.16 * len [m]) * speed [Gbps] + 2.72 MTU [B]) */ +/* xoff = ((301+2.16 * len [m]) * speed [Gbps] + 2.72 MTU [B]) + * minimum speed value is 40Gbps + */ static u32 calculate_xoff(struct mlx5e_priv *priv, unsigned int mtu) { u32 speed; @@ -130,10 +132,9 @@ static u32 calculate_xoff(struct mlx5e_priv *priv, unsigned int mtu) int err; err = mlx5e_port_linkspeed(priv->mdev, &speed); - if (err) { - mlx5_core_warn(priv->mdev, "cannot get port speed\n"); - return 0; - } + if (err) + speed = SPEED_40000; + speed = max_t(u32, speed, SPEED_40000); xoff = (301 + 216 * priv->dcbx.cable_len / 100) * speed / 1000 + 272 * mtu / 100; From e28408e98bced123038857b6e3c81fa12a2e3e68 Mon Sep 17 00:00:00 2001 From: Huy Nguyen Date: Thu, 7 Mar 2019 14:07:32 -0600 Subject: [PATCH 14/15] net/mlx5e: Update xon formula Set xon = xoff - netdev's max_mtu. netdev's max_mtu will give enough time for the pause frame to arrive at the sender. Fixes: 0696d60853d5 ("net/mlx5e: Receive buffer configuration") Signed-off-by: Huy Nguyen Signed-off-by: Saeed Mahameed --- .../mellanox/mlx5/core/en/port_buffer.c | 28 +++++++++++-------- 1 file changed, 16 insertions(+), 12 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c index f00de0c987cd6..4ab0d030b5448 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en/port_buffer.c @@ -143,7 +143,7 @@ static u32 calculate_xoff(struct mlx5e_priv *priv, unsigned int mtu) } static int update_xoff_threshold(struct mlx5e_port_buffer *port_buffer, - u32 xoff, unsigned int mtu) + u32 xoff, unsigned int max_mtu) { int i; @@ -155,11 +155,12 @@ static int update_xoff_threshold(struct mlx5e_port_buffer *port_buffer, } if (port_buffer->buffer[i].size < - (xoff + mtu + (1 << MLX5E_BUFFER_CELL_SHIFT))) + (xoff + max_mtu + (1 << MLX5E_BUFFER_CELL_SHIFT))) return -ENOMEM; port_buffer->buffer[i].xoff = port_buffer->buffer[i].size - xoff; - port_buffer->buffer[i].xon = port_buffer->buffer[i].xoff - mtu; + port_buffer->buffer[i].xon = + port_buffer->buffer[i].xoff - max_mtu; } return 0; @@ -167,7 +168,7 @@ static int update_xoff_threshold(struct mlx5e_port_buffer *port_buffer, /** * update_buffer_lossy() - * mtu: device's MTU + * max_mtu: netdev's max_mtu * pfc_en: current pfc configuration * buffer: current prio to buffer mapping * xoff: xoff value @@ -184,7 +185,7 @@ static int update_xoff_threshold(struct mlx5e_port_buffer *port_buffer, * Return 0 if no error. * Set change to true if buffer configuration is modified. */ -static int update_buffer_lossy(unsigned int mtu, +static int update_buffer_lossy(unsigned int max_mtu, u8 pfc_en, u8 *buffer, u32 xoff, struct mlx5e_port_buffer *port_buffer, bool *change) @@ -221,7 +222,7 @@ static int update_buffer_lossy(unsigned int mtu, } if (changed) { - err = update_xoff_threshold(port_buffer, xoff, mtu); + err = update_xoff_threshold(port_buffer, xoff, max_mtu); if (err) return err; @@ -231,6 +232,7 @@ static int update_buffer_lossy(unsigned int mtu, return 0; } +#define MINIMUM_MAX_MTU 9216 int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, u32 change, unsigned int mtu, struct ieee_pfc *pfc, @@ -242,12 +244,14 @@ int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, bool update_prio2buffer = false; u8 buffer[MLX5E_MAX_PRIORITY]; bool update_buffer = false; + unsigned int max_mtu; u32 total_used = 0; u8 curr_pfc_en; int err; int i; mlx5e_dbg(HW, priv, "%s: change=%x\n", __func__, change); + max_mtu = max_t(unsigned int, priv->netdev->max_mtu, MINIMUM_MAX_MTU); err = mlx5e_port_query_buffer(priv, &port_buffer); if (err) @@ -255,7 +259,7 @@ int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, if (change & MLX5E_PORT_BUFFER_CABLE_LEN) { update_buffer = true; - err = update_xoff_threshold(&port_buffer, xoff, mtu); + err = update_xoff_threshold(&port_buffer, xoff, max_mtu); if (err) return err; } @@ -265,7 +269,7 @@ int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, if (err) return err; - err = update_buffer_lossy(mtu, pfc->pfc_en, buffer, xoff, + err = update_buffer_lossy(max_mtu, pfc->pfc_en, buffer, xoff, &port_buffer, &update_buffer); if (err) return err; @@ -277,8 +281,8 @@ int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, if (err) return err; - err = update_buffer_lossy(mtu, curr_pfc_en, prio2buffer, xoff, - &port_buffer, &update_buffer); + err = update_buffer_lossy(max_mtu, curr_pfc_en, prio2buffer, + xoff, &port_buffer, &update_buffer); if (err) return err; } @@ -302,7 +306,7 @@ int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, return -EINVAL; update_buffer = true; - err = update_xoff_threshold(&port_buffer, xoff, mtu); + err = update_xoff_threshold(&port_buffer, xoff, max_mtu); if (err) return err; } @@ -310,7 +314,7 @@ int mlx5e_port_manual_buffer_config(struct mlx5e_priv *priv, /* Need to update buffer configuration if xoff value is changed */ if (!update_buffer && xoff != priv->dcbx.xoff) { update_buffer = true; - err = update_xoff_threshold(&port_buffer, xoff, mtu); + err = update_xoff_threshold(&port_buffer, xoff, max_mtu); if (err) return err; } From 7f1a546e322287ae948e0f5eb8d12b7b638d93a6 Mon Sep 17 00:00:00 2001 From: Eli Britstein Date: Mon, 18 Mar 2019 09:25:59 +0000 Subject: [PATCH 15/15] net/mlx5e: Consider tunnel type for encap contexts The driver allocates an encap context based on the tunnel properties, and reuse that context for all flows using the same tunnel properties. Commit df2ef3bff193 ("net/mlx5e: Add GRE protocol offloading") introduced another tunnel protocol other than the single VXLAN previously supported. A flow that uses a tunnel with the same tunnel properties but with a different tunnel type (GRE vs VXLAN for example) would mistakenly reuse the previous alocated context, causing the traffic to be sent with the wrong encapsulation. Fix that by considering the tunnel type for encap contexts. Fixes: df2ef3bff193 ("net/mlx5e: Add GRE protocol offloading") Signed-off-by: Eli Britstein Reviewed-by: Roi Dayan Signed-off-by: Saeed Mahameed --- .../net/ethernet/mellanox/mlx5/core/en_tc.c | 28 +++++++++++++------ 1 file changed, 19 insertions(+), 9 deletions(-) diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c index 5fb5cab36bf65..d75dc44eb2ff6 100644 --- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c @@ -2384,15 +2384,22 @@ static int parse_tc_nic_actions(struct mlx5e_priv *priv, return 0; } -static inline int cmp_encap_info(struct ip_tunnel_key *a, - struct ip_tunnel_key *b) +struct encap_key { + struct ip_tunnel_key *ip_tun_key; + int tunnel_type; +}; + +static inline int cmp_encap_info(struct encap_key *a, + struct encap_key *b) { - return memcmp(a, b, sizeof(*a)); + return memcmp(a->ip_tun_key, b->ip_tun_key, sizeof(*a->ip_tun_key)) || + a->tunnel_type != b->tunnel_type; } -static inline int hash_encap_info(struct ip_tunnel_key *key) +static inline int hash_encap_info(struct encap_key *key) { - return jhash(key, sizeof(*key), 0); + return jhash(key->ip_tun_key, sizeof(*key->ip_tun_key), + key->tunnel_type); } @@ -2423,7 +2430,7 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv, struct mlx5_esw_flow_attr *attr = flow->esw_attr; struct mlx5e_tc_flow_parse_attr *parse_attr; struct ip_tunnel_info *tun_info; - struct ip_tunnel_key *key; + struct encap_key key, e_key; struct mlx5e_encap_entry *e; unsigned short family; uintptr_t hash_key; @@ -2433,13 +2440,16 @@ static int mlx5e_attach_encap(struct mlx5e_priv *priv, parse_attr = attr->parse_attr; tun_info = &parse_attr->tun_info[out_index]; family = ip_tunnel_info_af(tun_info); - key = &tun_info->key; + key.ip_tun_key = &tun_info->key; + key.tunnel_type = mlx5e_tc_tun_get_type(mirred_dev); - hash_key = hash_encap_info(key); + hash_key = hash_encap_info(&key); hash_for_each_possible_rcu(esw->offloads.encap_tbl, e, encap_hlist, hash_key) { - if (!cmp_encap_info(&e->tun_info.key, key)) { + e_key.ip_tun_key = &e->tun_info.key; + e_key.tunnel_type = e->tunnel_type; + if (!cmp_encap_info(&e_key, &key)) { found = true; break; }